Re: [MiNT] Thing and MiNT

[Lonny Pursell <atari@bright.net>]

on 12/20/09 4:10 PM, Roger Burrows wrote:

> On 20 Dec 2009 at 21:35, Jo Even Skarstein wrote:
>> 
>> Peter Slegg skreiv:
>> 
>>> What you are saying is that an application that has worked
>>> fine for years with TOS and various versions of Mint is
>>> broken because some versions of XaAES or Mint break it.
>> 
>> No. I'm saying that Thing 1.27 most likely has been broken all the time.
>> A change (don't know which) in XaAES has highlighted a bug in Thing. The
>> fact that is has been working previously doesn't mean that it has been
>> working correctly.
>> 
> 
> To add to this:
> It would be difficult (and almost certainly a waste of resources) to determine
> whether Thing 1.27 is in fact broken.  But I think that most programmers have
> experienced the situation where a working program does not work with a new
> version of the operating system (or new hardware, or new something else) and
> have discovered that there was in fact a bug in the program which didn't show
> up before.  It would be unreasonable in this situation to expect that the
> operating system maintainers would work around a bug in another program ... in
> many cases, it might in fact be impossible to do so.
> 
> So when a new release of an operating system comes out, and a particular
> program no longer works properly (especially if it is only that program), it
> is 
> quite likely due to a program bug.  Ideally, you would contact the author, who
> would investigate, discover whether it's a program or operating system bug,
> and 
> either fix it or inform the operating system maintainers who would then fix
> their problem.  Unfortunately I don't think this is going to happen with
> Thing. 
> So you are faced with a (hard) choice:
>  . live with the bug (work around it)
>  . don't use the new release of the operating system
>  . stop using the program and find a replacement
> 
> This is one argument for open-source software - or for at least the publishing
> of source code for no-longer-maintained programs (although sometimes that
> source code might be unmaintainable anyway, based on my experience...).

There are a few smaller issues at the user level (non-dev).  Ok, v1.29b
solves the mysterious problems, but I can't deal with a German desktop.
Simply can't read it, I skipped the optional German classes in high school.
:-P  That said, I'm stuck with v1.27, therefore I'm very reluctant to change
out my old kernel or other components for that matter since no one seems
absolutely sure where the problem really is.

I've also lost my only copy of the 1.29b archive Thomas so graciously sent
me upon request. Like I fool I left it in the M$ mailer which eventually ate
it.  :-(  Ironically it's the only M$ app on my Mac and yet M$ and data loss
seem to go hand in hand.  Go figure.

-- 
Lonny Pursell    http://www.bright.net/~gfabasic/